paper-with-me

Papers

Generalizable Task Representation Learning for Offline Meta-Reinforcement Learning with Data Limitations

2023-12-26 · Renzhe Zhou, Chen-Xiao Gao, Zongzhang Zhang, Yang Yu

Generalization and sample efficiency have been long-standing issues concerning reinforcement learning, and thus the field of Offline Meta-Reinforcement Learning~(OMRL) has gained increasing attention due to its potential of solving a wide range of problems with static and limited offline data. Existing OMRL methods often assume sufficient training tasks and data coverage to apply contrastive learning to extract task representations. However, such assumptions are not applicable in several real-world applications and thus undermine the generalization ability of the representations. In this paper, we consider OMRL with two types of data limitations: limited training tasks and limited behavior diversity and propose a novel algorithm called GENTLE for learning generalizable task representations in the face of data limitations. GENTLE employs Task Auto-Encoder~(TAE), which is an encoder-decoder architecture to extract the characteristics of the tasks. Unlike existing methods, TAE is optimized solely by reconstruction of the state transition and reward, which captures the generative structure of the task models and produces generalizable representations when training tasks are limited. To alleviate the effect of limited behavior diversity, we consistently construct pseudo-transitions to align the data distribution used to train TAE with the data distribution encountered during testing. Empirically, GENTLE significantly outperforms existing OMRL methods on both in-distribution tasks and out-of-distribution tasks across both the given-context protocol and the one-shot protocol.

📄 PDF Abstract BibTeX arXiv:2312.15909

Code (1)

lamda-rl/gentle 공식 구현 pytorch

Tasks

Contrastive LearningDecoderDiversityMeta Reinforcement Learningreinforcement-learningReinforcement LearningRepresentation Learning

Methods 이 논문이 사용한 방법론

Contrastive Learning 설명 없음
ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Robust Task Representations for Offline Meta-Reinforcement Learning via Contrastive Learning

2022-06-21 · Haoqi Yuan, Zongqing Lu

We study offline meta-reinforcement learning, a practical reinforcement learning paradigm that learns from offline data to adapt to new tasks. The distribution of offline data is determined jointly by the behavior policy…

Contrastive LearningMeta Reinforcement Learningreinforcement-learningReinforcement Learning+2

Decoupling Meta-Reinforcement Learning with Gaussian Task Contexts and Skills

2023-12-11 · Hongcai He, Anjie Zhu, Shuang Liang, Feiyu Chen 외

Offline meta-reinforcement learning (meta-RL) methods, which adapt to unseen target tasks with prior experience, are essential in robot control tasks. Current methods typically utilize task contexts and skills as prior e…

continuous-controlContinuous ControlMeta Reinforcement LearningQuantization+3

Meta-Reinforcement Learning for Adaptive Control of Second Order Systems

2022-09-19 · Daniel G. McClement, Nathan P. Lawrence, Michael G. Forbes, Philip D. Loewen 외

Meta-learning is a branch of machine learning which aims to synthesize data from a distribution of related tasks to efficiently solve new ones. In process control, many systems have similar and well-understood dynamics, …

Meta-LearningMeta Reinforcement Learningreinforcement-learningReinforcement Learning+1

Meta-Reinforcement Learning for the Tuning of PI Controllers: An Offline Approach

2022-03-17 · Daniel G. McClement, Nathan P. Lawrence, Johan U. Backstrom, Philip D. Loewen 외

Meta-learning is a branch of machine learning which trains neural network models to synthesize a wide variety of data in order to rapidly solve new problems. In process control, many systems have similar and well-underst…

Meta-LearningMeta Reinforcement Learningreinforcement-learningReinforcement Learning (RL)

Your Offline Policy is Not Trustworthy: Bilevel Reinforcement Learning for Sequential Portfolio Optimization

2025-05-19 · Haochen Yuan, Minting Pan, Yunbo Wang, Siyu Gao 외

Reinforcement learning (RL) has shown significant promise for sequential portfolio optimization tasks, such as stock trading, where the objective is to maximize cumulative returns while minimizing risks using historical …

Offline RLPortfolio OptimizationReinforcement Learning (RL)Stock Prediction